English

Supervision and Source Domain Impact on Representation Learning: A Histopathology Case Study

Computer Vision and Pattern Recognition 2020-09-07 v1 Machine Learning

Abstract

As many algorithms depend on a suitable representation of data, learning unique features is considered a crucial task. Although supervised techniques using deep neural networks have boosted the performance of representation learning, the need for a large set of labeled data limits the application of such methods. As an example, high-quality delineations of regions of interest in the field of pathology is a tedious and time-consuming task due to the large image dimensions. In this work, we explored the performance of a deep neural network and triplet loss in the area of representation learning. We investigated the notion of similarity and dissimilarity in pathology whole-slide images and compared different setups from unsupervised and semi-supervised to supervised learning in our experiments. Additionally, different approaches were tested, applying few-shot learning on two publicly available pathology image datasets. We achieved high accuracy and generalization when the learned representations were applied to two different pathology datasets.

Keywords

Cite

@article{arxiv.2005.08629,
  title  = {Supervision and Source Domain Impact on Representation Learning: A Histopathology Case Study},
  author = {Milad Sikaroudi and Amir Safarpoor and Benyamin Ghojogh and Sobhan Shafiei and Mark Crowley and H. R. Tizhoosh},
  journal= {arXiv preprint arXiv:2005.08629},
  year   = {2020}
}

Comments

Accepted for presentation at the 42nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC'20)

R2 v1 2026-06-23T15:37:24.099Z